November 18, 2012
The Paradox of Voting
The Paradox of Voting--Posner
The paradox of voting in national elections is that, since a single vote is almost certain to have no effect on the outcome (in a Presidential election, it will merely add one digit to an eight-figure number), there seems to be no benefit from voting. The cost is small enough (if it's high for a person, he is unlikely to vote), but it's positive, so that if the benefit of voting is zero the voter is being irrational. Yet, as Becker points out, more than 100 million people bothered to vote in the recent Presidential election.
Some people vote because the political campaigns make it costly for them not to vote—one technique in "get out the vote" drives is pestering people to vote so that they will feel uncomfortable not voting. Some vote because they think that it will encourage others to do so. Some vote because they consider it a civic duty. Some voting is purely expressive—a way of expressing strong feelings pro or con a candidate (or pro one and con his opponent); certainly anger played a role in votes against Romney by members of groups that he or his party seemed to disrespect, and anger at Obama played a role in the large number of votes that Romney received. In this respect voting is like booing or cheering at an athletic event or other entertainment. One person's applause at a concert is inaudible to the performers, yet people applaud, and not mainly I think because others in the audience would look askance at them if they did not. And finally people interested in politics like to vote to convince themselves and others that their interest is serious—they are willing to put their money (not money exactly, but the cost in time and bother of voting) where their mouth is.
No one thinks that applauding is irrational, even though like voting it has no instrumental value, and has some, though very slight, cost.
Campaign advertising and other politicking—debates, the party conventions, etc.—try to excite voters about the election and induce them to vote for particular candidates, and in this respect closely resemble the advertising of consumer products and services. As Becker points out, much campaign rhetoric has little intellectual content, conveys little information, plays on the audience's emotions, and so on, but in these respects it doesn't differ much from the marketing of consumer products and services.
What is wasteful is not the time people spend voting, since they vote only if on balance it is a more valuable use of their time than the alternatives; what may be wasteful is the amount of money that is spent on electioneering, and the amount of candidate time spent on money raising. As with product advertising, so with political advertising, much of the expenditure is offsetting, so that if collusion were permitted there might be a net social benefit. (That is the rationale for campaign expenditure limits.) But there might not be, because of the consumption value of the political advertising and its contribution to the value that people derive from voting.
Why People Vote and What That Implies-Becker
The number of voters in the recent US presidential election was down by about 5 million voters from 2008, yet still about 120 million Americans voted this month. President Obama had one of the smallest winning percentages in the popular vote of incumbent presidents, but he still won by over 3 million voters.
This raises the very old question of why people vote in large elections when their chances of being a pivotal voter are virtually zero, and when voting takes time and is often inconvenient. The electorate is surely conscious of the cost to them of voting since, for example, turnout is usually much smaller when the weather is very bad. The common answer nowadays about this so-called paradox of voting is not that voters are irrational, but rather that they vote for reasons other than to influence outcomes. They may vote to indicate their moral support for particular candidates, or because they believe they express a precious right when they vote, or for other non-instrumental reasons.
These explanations do not say a lot without further analysis, but they do give possible directions for trying to understand the fundamental question of why and how people vote. Jorg Spenkuch, a graduate student at the University of Chicago, using data for German elections provides some answers. He shows that about 1/3 of the voters who prefer small parties, such as the Greens, acted "pivotally", and voted instead for parties that had much better chances of winning certain elections. If voters mainly want to express moral support for their favorite candidates, why do so many supporters of small parties that are very unlikely to win often vote for candidates of one of the big parties because they do not want to "throw away" their votes?
Since the great majority of voters presumably do not expect to influence election outcomes, who they support is influenced disproportionately by campaign rhetoric, debates among candidates that have little intellectual content, and by other methods of persuasion that are not very informative about the candidates. I like to say that consumers put more time and effort into deciding which cereal to buy and into other small consumer choices than in gathering information on economic and other issues about presidential candidates.
But I am not claiming that voters are less "rational" than consumers of everyday products. Individuals pay more attention to what they buy than whom they vote for because what they buy has a direct and tangible effect on their wellbeing. Since the incentives to become well informed are radically different, "rational" voting implies very different kinds of behavior than does rational choice of cereals or peanut butter.
Given the little effort that voters put into determining policy positions of different candidates, this improves the chances of candidates who are charismatic, look honest and caring, and project images that they have attractive visions of what they want to do even when they really have no well thought out beliefs about how to proceed if they get elected. This analysis, and the greater competition usually present in the private sector than in the political arena, imply that campaign rhetoric and campaign advertising will generally be less informative and less honest than is the typical private advertising. Of course, voters also value debates and campaign advertisements for the entertainment they provide (this is true of some advertising of goods as well).
Votes are more likely to matter in local contested elections, so one would expect voters to be better informed on local issues and local candidates than on more national issues and ational candidates. This is one reason why decentralized government is more likely to be responsive to the wishes of the electorate than is centralized government.
Of course, I agree that contested elections, the foundation of a democratic society, are far better than a system where there are no elections and a single Party decides who holds various government positions. Even weak competition among parties for political office with voters who are not well informed is usually far better than no competition at all. Winston Churchill famously said many years ago, "It has been said that democracy is the worst form of government except all those other forms that have been tried". This conclusion may not be quite as true in the modern Web world as it was in Churchill's time, but it still contains a fundamental truth.
Nevertheless, in determining which decisions should be made by the political process in a democracy, it is important to remember that individuals have little incentive to be well informed on political issues. As a result, their votes may not effectively reflect their interests or much knowledge of the issues.